Why must every remote call have a timeout, and how do you pick its value?

Answer

Without one, a hung dependency holds the caller's threads and connections forever, until the caller fails too, and the failure cascades upstream. Set it from the dependency's normal p99 latency plus a margin, and within the time the caller itself has left.

Review this in your daily deck → All cards in Resilience